Cool! I'd love to see Drill support for the Ibis DSL (which has semantic completeness with SQL):
http://docs.ibis-project.org/sql.html As Jacques alluded, we are working on a toolchain that will enable pandas-like UDFs that work natively with Drill and other SQL-like engines that support the ValueVector in-memory columnar protocol. This is quite a bit of work but one of my primary initiatives in 2016. cheers, Wes On Mon, Jan 11, 2016 at 7:34 AM, Wojciech Nowak <[email protected]> wrote: > Hello! > > You can try initial work of my python driver > https://github.com/PythonicNinja/pydrill > > Feel free to write your feedback. > > Based on that package I will create pydrill-dsl which will enable creating > queries without need to write sql. > https://gist.github.com/PythonicNinja/9b4952b6cbc17572c7db#file-pydrill_dsl-py > > -- > Wojciech Nowak > > W dniu 9 stycznia 2016 @ 20:34:04, Jacques Nadeau ([email protected]) > napisał(a): > > The easiest way to accomplish this today is to use Python ODBC bindings and > the Drill ODBC driver. Wes McKinney (cc'd, of Cloudera, Pandas and Ibis) > has started work on some native Python bindings for the Drill in-memory > format and once his work is a little way along, we should be able to build > on that to provide a nice set of native Python bindings that would be even > easier to use. > > -- > Jacques Nadeau > CTO and Co-Founder, Dremio > > On Fri, Jan 8, 2016 at 3:55 PM, Jin Chen <[email protected]> wrote: > >> Hi, >> Just want to know whether apachdrill has plan to support client binding in >> the future. >> Such as python/nodejs/golang client, so I could write program to query >> automatically. >> >> Thanks, >> Jin >>
